darwin {robustmatrix}

DARWIN (Diagnosis AlzheimeR WIth haNdwriting)

Description

The DARWIN (Diagnosis AlzheimeR WIth haNdwriting) dataset comprises handwriting samples from 174 individuals. Among them, 89 have been diagnosed with Alzheimer's disease (AD), while the remaining 85 are considered healthy subjects (H). Each participant completed 25 handwriting tasks on paper, and their pen movements were recorded using a graphic tablet. From the raw handwriting data, a set of 18 features was extracted.

Usage

data(darwin)

Format

An array of dimension (p,q,n), comprising n = 174 observations, each represented by a p = 18 times q = 25 dimensional matrix. The observed parameters are:

• Total Time

• Air Time

• Paper Time

• Mean Speed on paper

• Mean Acceleration on paper

• Mean Acceleration in air

• Mean Jerk on paper

• Pressure Mean

• Pressure Variance

• Generalization of the Mean Relative Tremor (GMRT) on paper

• GMTR in air

• Mean GMRT

• Pendowns Number

• Max X Extension

• Max Y Extension

• Dispersion Index

Source

UC Irvine Machine Learning Repository - DARWIN - doi:10.24432/C55D0K

References

Cilia ND, De Stefano C, Fontanella F, Di Freca AS (2018). “An experimental protocol to support cognitive impairment diagnosis by using handwriting analysis.” Procedia Computer Science, 141, 466–471.
Cilia ND, De Gregorio G, De Stefano C, Fontanella F, Marcelli A, Parziale A (2022). “Diagnosing Alzheimer’s disease from on-line handwriting: a novel dataset and performance benchmarking.” Engineering Applications of Artificial Intelligence, 111, 104822.